The plot itself
An HPLC chromatogram is a one-dimensional record of what came out of a column and when. The horizontal axis is time, in minutes. The vertical axis is absorbance at a chosen wavelength — 220 nm is the usual pick for peptide backbones, because the amide bond absorbs reliably there. Every peak above the baseline is a compound the column separated out at a given retention time.
The column itself is almost always a reverse-phase C18. The peptide is injected in a low-organic solvent, and a gradient of acetonitrile (with trifluoroacetic acid as an ion pairing agent) pulls compounds off the stationary phase in order of hydrophobicity. The most polar species elute first; the most lipophilic last. This is not magic — it’s just partitioning.
A single purity percentage is a conclusion. A chromatogram is the evidence. Don’t accept the conclusion without the evidence.
Where the 99% comes from
The purity percentage is the area under the main peak, divided by the total area of all peaks the integrator identified, multiplied by 100. It is a ratio, not a measurement of mass. That matters: a peptide can be 99.4% of the UV-absorbing material and still only be 80% of what’s in the vial by mass. The other 20% might be water, salt, or counter-ion.
This is why a serious COA publishes both the RP-HPLC purity and the peptide content by amino acid analysis (AAA). If the vendor shows you one and not the other, you are seeing half the story.
What a “clean” chromatogram looks like
A healthy trace has:
- A flat baseline before the main peak — not drifting upward, not wavy.
- A tall, symmetric main peak, narrow enough that it doesn’t shoulder into neighbors.
- Minor peaks accounted for and quantified, not hand-waved away as “solvent front” when they clearly elute at relevant retention times.
- An integration start-and-stop that is visible and sensible — not starting at 4.2 min because that’s where the interesting story begins.
Three tells of a bad report
The integrator is cropping the range. If you see a chromatogram that runs from 4.0 to 8.0 min, ask what happened to 0 – 4.0 and 8.0 – end. Hydrophilic impurities elute early; hydrophobic ones elute late. Cropping hides both.
The baseline is clearly drifting. Gradient HPLC can produce a slightly rising baseline in the late part of the run — that’s normal. A baseline that drifts in the flat region before the main peak usually means the column is dirty, the injection was contaminated, or the detector is flaky.
The main peak is asymmetric and tailing. Asymmetric peaks are often real; tailing peaks usually mean the sample has secondary retention with the silica, or the pH is wrong, or the column is overloaded. If the vendor’s stated purity depends on integrating a heavily tailed peak, ask for the Assymetry factor.
Retention time is not identity
A peak that lands at the expected retention time of your peptide is consistent with your peptide being there. It is not proof. That’s what the LC-MS is for: it tells you the mass of what eluted, which combined with the retention time gives you a much higher-confidence identification. We insist on both.
What to ask for
For any research peptide you intend to use, request: the full chromatogram (not a cropped view), the peptide content by AAA, the LC-MS identity confirmation of the main peak, and the endotoxin result. If the supplier cannot show you all four, their confidence in the lot is lower than yours is.